In the perception of Siemens technicians, industry 4.0, “... also called the Fourth Industrial Revolution, is a concept, presented to the public for the first time at the 2011 Hannover Fair, which represents industrial automation and the integration of technological solutions. innovative technologies such as artificial intelligence, internet of things, robotics and the use of cloud data. On the occasion of its name, technology companies, universities and research centers supported by the German government proposed a new vision for the operation of factories: a model with decentralized control of production processes and collaboration between intelligent devices. All of this is interconnected throughout the entire production and logistics chain, transforming the way machines communicate and use information to optimize production (1)”.

Nowadays, 2022, technologies like Blockchain, Augmented Reality, Virtual Reality, Metaverse, Big Data and Analytics, 3D Printing etc. are highlights within the scope of Industry 4.0, joining Artificial Intelligence in its various modalities, the Internet of Things, etc.

Equally relevant are issues related to digital infrastructure: hybrid cloud computing and cyber-security issues in the broadest possible sense.

Technological Innovation in Business

The World Economic Forum

According to experts at the World Economic Forum, “The Fourth Industrial Revolution represents a fundamental change in the way we live and work. It is a new chapter in human development, made possible by advances that are commensurate with those of the first, second and third industrial revolutions - merging the physical, digital and biological worlds and fusing technologies in ways that create both promise and danger. The speed, breadth and depth of this revolution forced us to rethink how countries should develop, how organizations create value, and how people from all walks of life can benefit from innovation. Now, as the world grapples with COVID-19, there is an opportunity to further embrace this revolution in ways that create a more inclusive and human-centered global economy” (3).

The complexity resulting from the integrated use of Blockchain, Internet of Things and Artificial Intelligence technologies

The construction of ecosystems integrated by Blockchain architecture can become reasonably complex if we consider the subsystems for capturing data via IoT tools, as in the case of industrial production in which processes depend on materials kept at controlled temperatures (in the case of the production of dairy products). , for example). Sensors detect temperatures outside the proper limits and send an alarm to the system. Actuators, after analyzing the situation by “smart contracts”, divert the production batch outside the authorized parameters to the container of materials to be analyzed and eventually discarded. The records related to the occurrences will remain, reliably and traceable, in a Blockchain environment for future analysis.

The complexity can grow even more if the information to be recorded by the Blockchain happens as a result of decisions made with the use of Artificial Intelligence tools.

It should be noted that both the topic of computer vision and machine learning or even natural language processing, relevant topics in AI, need to have their logic explained in a clear and transparent way, which is not always the case, especially if we use networks. deep learning neural networks.

On the issue of explainability in the case of using Artificial Intelligence tools in ecosystems integrated by Blockchain.

“Explainable artificial intelligence (XAI)” is a set of processes and methods that allow human users to understand and trust the results and outputs created by machine learning algorithms.

Artificial Intelligence explainability (XAI) is used to describe an AI model, its expected impact and possible biases. It helps to characterize model accuracy, fairness, transparency, and results in AI-powered decision making. Explainable AI is crucial for an organization to build trust when putting AI models into production. AI explainability also helps an organization take a responsible approach to AI development. As AI becomes more advanced, humans are challenged to understand and redo how the algorithm arrived at a result. The entire calculation process is transformed into what is commonly called a "black box" that is impossible to interpret. These black box models are created directly from the data. And not even the engineers or data scientists who create the algorithm can understand or explain what exactly is going on inside them or how the AI ​​algorithm arrived at a specific result.” (4).

The integrated use of AI tools together with the processing of data from IoT devices, being processed in an ecosystem with several participants, from different categories and from different organizations (integrated via Blockchain) can provide an essential benefit to the issue of explainability in AI (XAI).

In short, the set of parameters and data audited in relation to the algorithm that made the decision will be able to register on the Blockchain.

Suppose, for example, a decision to withdraw an operator from an industrial assembly line as a result of the decision-making of an algorithm that considered the practices of the worker in question to be reckless. Removed from his duties, the worker may question the company's management regarding the reasons for the decision making that removed him from the factory's operation. Using the AI ​​Explainability (XAI) subsystem, registration is carried out in the Blockchain environment. Thus, it will be possible to track the decision made by AI, with a guarantee of the veracity of the information.

Ante Mesin (5) recalls that Blockchain architecture, also known as “Distributed Ledger Technology (DLT)” and artificial intelligence (AI), when used together, allow machines to perform intelligent tasks while blockchain technology records and stores data. and events chronologically. We know that the blockchain records the data deposited in it in an immutable and traceable way, also enabling the integration of closed subsystems within their organizations and open in relation to the data relevant to the ecosystem transactions. This is the case with Renault, which, using blockchain, integrates participants from different categories (parts suppliers, certification teams, industrial maintenance service providers, customers, etc.) the internal data of each of the categories of participants is protected (6).

It is up to the architects of a system with these characteristics to design the project, including the issue of explainability in light of the ethics that govern the business in which the aforementioned technologies are inserted.